AWS: Amazon Bio Discovery
AWS announced Amazon Bio Discovery, a service that combines AI with wet‑lab integration to reduce antibody design timelines from about a year to weeks, and the announcement cites a pilot with Memorial Sloan Kettering. The service is framed as accelerating biologics design through tighter AI‑lab workflows. (x.com)
Amazon Web Services has launched Amazon Bio Discovery, a new software service that lets drug researchers use artificial intelligence models and send the best molecule designs straight into lab testing. (aboutamazon.com) The service was announced on April 14, 2026, at the AWS Life Sciences Symposium in New York. Amazon Web Services said the product gives scientists access to more than 40 biology models and uses software agents to help choose models, tune settings, and rank candidates for experiments. (aws.amazon.com 1) (aws.amazon.com 2) Drug discovery starts with finding molecules that can bind to a biological target, and antibodies are one common class of those molecules. In this case, Amazon Web Services is pitching a system that links computer design work with contract research organizations that run the physical lab tests and return the results to the software for the next round. (aws.amazon.com) Amazon Web Services said a pilot with Memorial Sloan Kettering Cancer Center used the system to design nearly 300,000 novel antibody molecules, narrow them to 100,000 candidates, and move them into wet-lab testing in weeks rather than up to a year. The company said the work focused on potential pediatric cancer therapies. (aws.amazon.com) (aboutamazon.com) Memorial Sloan Kettering and Amazon Web Services had already announced a broader cancer research collaboration in November 2024. That earlier deal covered artificial intelligence, high performance computing, and cloud tools for cancer research, and the new product turns part of that work into a commercial Amazon Web Services offering. (mskcc.org) (aboutamazon.com) The company’s own case study gives more detail on the pilot. Amazon Web Services said researchers generated 288,000 nanobody designs across eight target regions, cut that set to 100,000 for yeast-display screening, and got reliable binding measurements for 46 of 116 tested candidates, with affinities ranging from 0.66 nanomolar to 305 nanomolar. (aws.amazon.com) Amazon Web Services is not offering a claim of full drug development in weeks. Reuters, citing Amazon Web Services executive Dan Sheeran, reported that the system is meant to speed early-stage research and “augment, not replace” scientists and contract research organizations. (usnews.com) The launch also puts Amazon Web Services more directly into a crowded market for artificial intelligence drug-discovery tools. Axios reported that the move intensifies competition among technology and biotech companies trying to lower the cost and time of early research, while Amazon Web Services is tying its pitch to cloud infrastructure and model access rather than a single in-house drug pipeline. (axios.com)
